Mind the Gap: A survey of chronic wound datasets, methods, and clinical translation
Chronic wounds such as diabetic foot ulcers, pressure ulcers and leg ulcers are a growing health problem worldwide, causing serious illness, disability and high costs. Computer programs that use artificial intelligence (AI) to analyse photographs of wounds could help doctors measure, track and assess wounds more consistently, and smartphone apps built on this technology are starting to appear. This survey looked at how well the research behind these tools matches what real-world apps claim to do, by reviewing 20 wound image datasets (14 public or partly public, 6 private), 50 AI methods, and 13 mobile wound-assessment apps, including both commercial clinical products and research prototypes. The authors found a serious mismatch: public image collections were mostly built to test and compare algorithms rather than to support real clinical use, so they often lack standardised photo-taking instructions, objects placed in the photo to show true size, depth information, detailed labelling of the different tissues in a wound, and images from people with a range of skin tones. At the same time, the apps advertise features such as automatic tissue classification, calibrated measurement and depth estimation for which there is not enough public data to prove they work reliably across patients. The paper maps what data exist against what apps claim, describes unresolved technical problems, and proposes practical recommendations for building wound-imaging systems that can actually be used in clinics. For patients, this means wound-tracking apps are promising but not yet fully proven; photo-based wound monitoring should support, not replace, hands-on assessment by a clinician. Limitations include that this is a survey of published material, not a new patient study, and app capabilities were taken from their documentation rather than tested head-to-head.
Key takeaways
- Wound-photo AI apps are promising, but the image databases used to develop them were not designed for real clinical settings or diverse patients.
- Reviewing 20 datasets, 50 AI methods and 13 apps, researchers found major gaps between research data and apps' advertised features.
- Public wound-image datasets often lack size markers, depth data, detailed tissue labels and images across different skin tones.
- Some apps claim automatic tissue classification, calibrated measurement and depth estimation that current public data cannot fully validate.
- For patients: photo-based wound tracking may help monitor healing, but should not replace in-person assessment by a wound-care clinician.
- Ask your doctor whether any app used on your wound has been validated, how it accounts for skin tone, and who reviews the results.
Read the full paper / source →
This content summarizes published medical research for educational purposes only. It does not constitute medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider before making treatment decisions.
Free consultation
Could this treatment help you?
Our DFU specialists review every inquiry personally. Tell us about your situation — response within 48 hours, no commitment required.
This website provides medical-information science popularization only. It does NOT provide online diagnosis, prescriptions, or medical advice. Consult a licensed physician for any medical decision.
Editor's note
Think of these wound apps as a smart measuring tape plus a trained eye in your phone: they photograph your foot ulcer and try to estimate its size, depth and tissue — like a GPS that tells you whether a wound is shrinking or spreading. If you've been told your ulcer won't heal or that amputation is the only option, this research doesn't offer a new cure — it shows that phone-based monitoring is getting better at helping clinicians spot problems early and track progress, which can support decisions about when to change treatment or escalate care. At China's specialist diabetic foot centres, this kind of AI wound assessment is not yet a routine part of standard care; it is mainly used in research settings, so an overseas patient should expect expert in-person wound assessment (debridement, offloading, infection control, vascular checks) as the mainstay, with imaging tools as a helpful extra — not the deciding factor.